Objectivist and Popperian Epistemology

Anonymous's picture
Submitted by Anonymous on Wed, 2013-07-03 17:14

Ayn Rand has the best moral philosophy ever invented. Karl Popper has the most important breakthrough in epistemology. Most Objectivists seem to think that Popper and Rand are incompatible, and Popper is an enemy of reason. They have not understood him. These lists are intended to help explain my motivation for integrating Rand and Popper, and also to help highlight many similarities they already have.

Points Popperian epistemology and Objectivist epistemology have in common. In Popperian epistemology I include additions and improvements by David Deutsch and myself:

- opposition to subjectivism and relativism
- fallibilism
- says that objective knowledge is attainable (in practice by fallible humans)
- realism: says reality is objective
- connected to reality: we have to observe reality, keep our ideas connected to reality
- asserts there is objective truth
- attention to context ("problem situation" or sometimes "problem" is the common Popperian term meaning context. E.g. a Popperian will ask "What is the problem this is addressing?" and be asking about context.)
- pro-science
- opposition to positivism
- opposition to the language analysis school of philosophy
- say that most professional philosophers are rather crap
- opposition to both skeptical and authoritarian schools of epistemology
- keeps our concepts "open-end[ed]" (ITOE). That means: possible to improve in the future as we learn more.
- says that there are objective moral truths
- does not seek a "frozen, arrested state of knowledge" (ITOE)
- written clearly and understandably, unlike much philosophy
- says epistemology is useful and valuable to real people; it matters to life; it's practical
- you can't force an idea on someone. they can choose to accept it or not
- you can't implant an idea in someone. you can't pour it in, stick it in with surgery, make them absorb it, etc. they get to think, interpret, choose.
- free will
- people are not born with some unchangeable nature and innate ideas. we can be self-made men. we can learn, change, improve, progress
- emphasis on active use of one's mind, active learning
- no inherent conflicts due to objective truth
- understanding of unconscious and inexplicit ideas
- if two ideas contradict, at least one is false
- integration of epistemology with morality, politics, and more
- rejection of authority
- full rejection of idealism, solipsism
- strong emphasis on clarity
- rejection of limits on human minds
- reject probabilistic approaches to epistemology
- looks at man as rational and capable
- value of critical thinking including self-criticism

Strengths of Objectivist epistemology:
- stolen concept
- package deal
- check your premises
- ideas about integrating all one's knowledge and removing all contradictions
- measurement omission and concept formation ideas both worthwhile, though flawed
- good criticisms of many opponents of reason
- good understanding of essentials vs non-essentials, e.g. for definitions
- idea about automating some thinking
- good explanation of what objectivity is
- Judge, and be prepared to be judged

Strengths of Popperian epistemology:
- evolution creates knowledge
- conjectures and refutations method
- piecemeal, incremental method. value of every little improvement
- identification of, and solution to, justificationism
- addresses induction
- conjectural, fallible, objective knowledge
- idea that we progress from misconception to better misconception
- myth of the framework
- value of culture clash
- emphasis on bold highly-criticizable claims, sticking your neck out to learn more
- no shame in mistakes
- value of criticism. criticism is a gift
- understanding of rationality as being about error correction
- unimportance of starting points. you can start anywhere, improve from there
- criticism of definitions
- criticism of foundations, bases
- criticism of essentialism
- criticism of manifest truth (and self-evidence, obviousness, etc)
- static and dynamic memes
- structural epistemology
- coercion and common preferences
- understanding of conflict and symmetry
- applications to parenting, education, relationships
- understanding of tradition
- explanation of value of external criticism (if everyone has some blind spots, but some people have different blind spots then each other, then it's productive to share criticism with each other. a little like comparative advantage)
- emphasis on critical method, criticism (ideas stand unless refuted)
- let our ideas die in our stead

Some of you are now wondering about details. I know. But it's so much! Let's do it like this: if you are interested in one of the topics, ask about it and I can elaborate. If you would preference a reference to existing material on the topic, that's fine too.

( categories: )

Alan wrote an answer to

curi's picture

Alan wrote an answer to Dykes. Enjoy.


"If I spend the time to

Tom Burroughes's picture

"If I spend the time to convince you this is completely false, what will I get? Why should I bother? It's such a bad, unserious and incompetent treatment that it's not much fun to deal with. Among various problems, Dykes simply does not understand Popper's most important ideas and isn't really part of the conversation."

You wanted to convince me of X, I was unconvinced, and you get ratty. Cool it down. I disagree with you - no need to take it personally. This is not a competition like a game of tennis. (And I repeat the point that I find a lot of your points interesting and said so).

And Dykes - is not playing silly semantic tricks here to trap or catch KP out. Popper was a man who told us to take precise use of language very seriously (and good for him). It strikes me as entirely legitimate to raise the point in the way Dykes did. This leads me to a broader point in that in much of our ordinary, normal life, we act as if induction is true, and don't necessarily use a falsifiability route to it.

Sorry about the reverence point: I got the impression here of you being overly sensitive.

As for your general assessment of Dykes' piece, one can bridle at the perhaps caustic tone of it, but given the length of the article and the sheer range of citations and quotations, it is unjust to say it is not a serious treatment and a hit-job. Dykes disagrees with the guy in many ways. That does not mean he was being flippant.

"This essay of Dykes is a

curi's picture

"This essay of Dykes is a serious treatment and a call for Popper"

If I spend the time to convince you this is completely false, what will I get? Why should I bother? It's such a bad, unserious and incompetent treatment that it's not much fun to deal with. Among various problems, Dykes simply does not understand Popper's most important ideas and isn't really part of the conversation.

The belief criticism is not arguing about any substantive issue so I don't see the point. It's not clear what the point is from what you quoted, but it's a well known claim that Popper said "I am not a belief philosopher" (paraphrase) and also used the word "belief" in a number of places, and that's supposedly bad. In English, words have more than one way of being used and Popper uses them in different ways in different places. Who cares?

The claim about reverence is particularly unfair. I have a long history of criticizing Popper, I've never been reverent, and I don't know of any Popperians (not even stupid purported ones) who are reverent.

Some of Popper's stuff is truly awful btw. E.g. in a letter to Magee he suggested the government have a 51% share in companies. His views on censorship and TV are terrible too (and seriously contradict both his epistemology and much of his other political thinking).

(BTW in light of stuff like this, the claim that Popper's thinking had inconsistencies is not something I regard as important at all. Everyone who studies Popper well already knows that. Who cares? The important thing is primarily his major breakthroughs in epistemology. Those can be true even if he's inconsistent in various ways.)

Dykes' paper

Tom Burroughes's picture

I should add that I have re-read Dykes' essay http://www.libertarian.co.uk/l... and he quotes Popper at length, is very detailed on his sources, the origins of some of the points, and I think your treatment of Dykes is still unfair. This essay of Dykes is a serious treatment and a call for Popper to be subjected to the same level of rigour that he sought, or claimed, to apply to other thinkers.

For instance, here is Dykes on Popper's inconsistency:

Yet one cannot help being struck when reading Popper by the extent to which he relied on what he denied. In
The Poverty of Historicism , for example, the words ‘believe’ or ‘belief’ often occur several times a page, viz: “I
believe that theories are prior to observations.... I do not believe , therefore, in the ‘method of generalization’ .... I
believe , rather, that the function of observation and experiment is the more modest one of help-ing us to test our theories...
”[POH 98]. “I do not believe that we ever make inductive generalisations.... I believe that the prejudice that we proceed in this way is a kind of optical illusion.... Now all this, I believe , is not only true for the natural but also for the social sciences...
”[POH 134-5]. In Objective Knowledge we read “I believe in the reality of the physical world. Secondly, I believe that the world of theoretical entities is real..” [OKN 323n7]. In Unended Quest , we are told of Popper's “conviction that there is a real world” and that he “became convinced that... we cannot start from our sense experiences” [UNQ 75]. In The Self and Its Brain we read:
“I wish to state clearly and unambiguously that I am convinced that selves exist” [UNQ 101]

In other words, Dykes goes to a lot of trouble to establish this inconsistency with widespread citations, to an almost painful degree. (I suspect that is why some Popperians get upset with Dykes - he is failing to show due reverence.)

Anyway, I am going to re-read some of my old KP books and see where that takes me.

"Dykes is right that a

Tom Burroughes's picture

"Dykes is right that a "conjecture" isn't knowledge. But "conjectural knowledge" is not the same thing as "a conjecture"."

Not sure I understand the point. To know something is different from guessing it or or making a conjecture, so I conjectural knowledge strikes me as an oxymoron. I "know" that I have made a conjecture, but I don't "know" that the conjecture is going to be valuable, or correct in some important way. That is why I think Dykes is right to draw attention to this point.

I am sure it makes sense if one is going to criticise or debate an author to read the author's books. I have read several of KP's, the first I remember reading was the Poverty of Historicism, a great book that ought to be more widely read than it is.


"I think the claim that there

curi's picture

"I think the claim that there is an asymmetry between confirmation and falsification is fair one" -- to be clear, this is not my own claim or idea. it's a particularly well known one of popper's ideas. that dykes doesn't seem to be aware of...

Regarding the quote about conjectural knowledge, this again goes under "Dykes doesn't know jack about Popper". he doesn't know what conjectural knowledge is. that's what the objection amounts to: he doesn't understand. (plus perhaps an element of unwillingness to question his premises about what knowledge is, a topic which certainly should be open to the possibility of debate).

Popper's big breakthrough in epistemology came by questioning some premises that had been pretty universally accepted for 2000 years. Dykes is unaware of this and doesn't discuss it.

Dykes is right that a "conjecture" isn't knowledge. But "conjectural knowledge" is not the same thing as "a conjecture". it's a different thing. Popper could just have well have called it "fallible knowledge". Fallibility is what he was trying to emphasizes, as against the standard Justified True Belief theory of knowledge which is infallibilist.

Would Dykes then be satisfied after a minor terminology change? (I actually rarely use the term "conjectural knowledge" btw. usually i prefer the terminology "knowledge"). if a minor terminology change would fix it then his objection is silly. if not, well, his argument about what a "conjecture" is doesn't explain a problem with the actual terminology-independent idea of popper's.

Objectivism has a rather ambiguous relationship with fallibility. Sometimes it says of course people are fallible. Then i hear Rand in a Q&A saying that Frank infallibly knew which music and art to get her that she'd like, but no one else was good at it. I don't know how you argue with that, she has no actual arguments it's true, it's ridiculous (Frank was very good at it, but not infallible), and yet she said it. that's just one example of many. a more important issue is the self-evident axioms which are treated as infallible -- meaning that you don't get to question and debate them because they are deemed "self-evident" which silences discussion. (and reminds me of Rand answering another Q&A with basically "if you don't understand that part of atlas shrugged, i'm not going to tell you". i like most of what Rand says, but there's the occasional thing i pretty strongly disagree with)

anyway you can't really judge what dykes understands by his bibliography (which FYI is missing important stuff like The Myth of the Framework). the approach i'm using is to have seriously studied the material, gotten to the point of understanding it very well according to the Popperian community (rather than just read a few books and maybe talk to some Objectivists and leave my knowledge uncriticized, unrefined, and actually low quality), and then compare what popper is about to what dykes talks about. they aren't even close.

take, "However much Popper may have rejected induction, his own method was in fact dependent upon it.". this is pretty much the equivalent of writing, "However much Rand may have rejected altruism, her own method was in fact dependent on it." It's seriously on that level.

really the only thing to do with stuff like Dykes is to explain popper to him from the beginning. except he's not listening. and anyway that's already been done: popper's books explain his ideas from the beginning. so what should i do? rewrite popper's books? what would i change? the problem isn't with the books but with how dykes approaches them. there is no way to write books that will work on any arbitrarily hostile audience. for books to explain big new ideas successfully requires people want to learn, make an effort, etc... the reader must take interest and initiative. there's nothing i can do about that. anyone who cares to learn popper can do so without me writing something new, and people who don't want to won't. so i don't really see what to do about this situation.

i admit popper is difficult to understand. (so is Rand, FYI. most people who read some Rand books do not understand Objectivism.) important ideas that substantially differ from one's culture are hard for most people to grasp. most people who read some popper books won't get it. but if i rewrote his material into a few books the situation wouldn't change, still most people reading a few of my books wouldn't get it. i think what people seriously interested in learning need to do is engage with people who already understand the material. ask questions as you read through the books, see if the people who know the material think your'e understanding it correctly or can offer corrections. i think discussion is the key. and i have made it available.

anyone who wants to understand popper can ask questions here: http://groups.yahoo.com/group/...

this discussion group (which I made) has the best popperians who are willing to have discussions. (btw another valuable thing besides asking questions and stating one's understanding of topics for criticism is to see how popperians actually use the ideas in their own discussions). it's also very Objectivism friendly (unlike Objectivist forums which are all Popper hostile).

if you want to understand popper and epistemology, i don't think you have much choice but to read primary sources yourself and participate in discussion. you'll have to make a serious effort. if you're unwilling, i don't think pointing out several more of Dykes' glaring errors would help anything.

Curi, before I respond to

Tom Burroughes's picture

Curi, before I respond to your response to me about what Dykes writes, I think it fair to say that you deserve great credit for trying to show that Objectivists can and should learn from the likes of Popper. It is one hell of a list you have presented there. It is going to be a good debate.

However, I think you are mistaken in claiming that Dykes "apparently never read much Popper". How can you prove such a negative? From what I see from Dykes' footnotes and bibliography used in his papers, he has read quite a lot of KP. I think you should not try to knock him down by asserting that someone has not read much of X. That's not playing fair.

I think the claim that there is an asymmetry between confirmation and falsification is fair one, but as Dykes points out, there is a lot of difference between falsifying a theory that we can be pretty certain was silly badly constructed, and one that was not, while there is a lot of difference between confirming X when X is a ramshackle idea based on little prior proof or theory, and confirming a solid theory, and so on. As Popper himself said, I think, we learn the most when a really solid-looking idea is falsified; if we falsify a theory about the existence of spaghetti monsters, not so much.

There is another point that Dykes makes which is worth quoting on the issue of what knowledge is, which I think is pretty powerful:

"Furthermore, the proposition ‘knowledge is conjectural’ is irrelevant. Epistemology, the branch of philosophy to which the theory of fallibilism belongs, is the study of the methods and grounds of knowledge. Conjecture is not valueless in this endeavour, but it is by definition not knowledge. Conjecture is conjecture, knowledge is something else entirely. The study of conjecture might be useful, but it is not germane to the study of knowledge. Another immediate objection is that the notion of ‘conjecture’ actually depends for its intelligibility upon the prior concept of ‘knowledge’. The idea of a ‘conjecture’ arose precisely to designate a form of mental activity which was unlike knowledge, and to distinguish clearly from knowledge an idea put forward as opinion without proof. This common error is called ‘the fallacy of the stolen concept’ in the Objectivist philosophy."

Back to induction, here is a good review of the recent David Harriman book on the subject.


He further apparently never read much Popper, b/c he doesn't know the asymmetry between confirmation and falsification. When you confirm something, it might still be false, even if your confirmation has no mistakes. Nothing definite happens. When you falsify, then if your falsification has no mistakes then the theory is false, and that's it. There's a definite result. Confirmation runs into questions like "How much confirmation should we get before we stop?" but falsification doesn't.


Kyrel Zantonavitch's picture

Curi -- Certainly an interesting series of claims! Smiling

Evidently I need to reexamine Karl Popper...

 I think Rand never claimed

curi's picture

I think Rand never claimed she had solved the problem of induction - she is not given credit for being honest on this.

I know it. I have given her credit. She says this in ITOE 2nd edition. I just quoted it yesterday and have done so before.

The basic problem with the Dykes piece is that he doesn't understand what Popper's positions are. From two things I wrote previously:


Popper built his philosophy on foundations borrowed from Hume and Kant. His first premise was wholehearted acceptance of Hume's attack on induction. 


Hume stated, in essence, that since all ideas are derived from experience we cannot have any valid ideas about future events - which have yet to be experienced. He therefore denied that the past can give us any information about the future. He further denied that there is any necessary connection between cause and effect.

Popper did not take the position that the past cannot give us any information about the future.


This is a poor attempt at criticizing Popper. The author doesn't know the difference between

A) uncertain

B) absolutely useless

He says:

However, recalling Popper's Kantian premise, one might reasonably enquire at this point: if all observations are theory-laden, and thereby suspect, what justifies our placing any confidence in negative observations? The procedure of observation is identical whether one is seeking evidence in favour of a theory, or testing for evidence against it. If our senses are automatically suspect, as Popper maintained, negative or falsifying instances deserve no more credibility than positive or confirming ones.

He takes 'our senses don't provide certainty' to mean 'all sense data should be rejected as false'. It's certain truth, or it's false, and that's it.

He further apparently never read much Popper, b/c he doesn't know the asymmetry between confirmation and falsification. When you confirm something, it might still be false, even if your confirmation has no mistakes. Nothing definite happens. When you falsify, then if your falsification has no mistakes then the theory is false, and that's it. There's a definite result. Confirmation runs into questions like "How much confirmation should we get before we stop?" but falsification doesn't.

I think Rand never claimed

Tom Burroughes's picture

I think Rand never claimed she had solved the problem of induction - she is not given credit for being honest on this. I am not entirely sure that Popper fully solved it though.

Here is an interesting treatment of Popper by Objectivist Nicholas Dykes, some time ago. He's quite scathing about Popper in some respects:


Rand's concept formation has

curi's picture

Rand's concept formation has elements of induction and doesn't emphasize the value of criticism.

Popper is rather different because he doesn't try to provide much in the way of guidelines for how positively to think. Instead he focusses more on how to correct errors in ideas. Think however you want (it's not all equal or arbitrary, but there's no strict rules you must follow), and then we'll judge and criticize (and try to improve) the results and conclusions, judging only by their content (not source).

There is no royal road to truth. There is no method that you use it to come up with ideas and your ideas are going to be awesome. No matter what you do, mistakes are going to happen and criticism and error correction are crucial.

One particular place Popper's epistemology is better is induction. Popper solved the problem of induction while Rand did not.

Mr Temple

Lindsay Perigo's picture

Captivating piece. Thanks for posting. I'm curious: what are the flaws you see in Rand's view of concept formation? I ask because in my view this is the area where she blows Popper out of the water, while you say his epistemology was better than hers. Please elaborate. As best I can tell, where he agrees with her, he wouldn't be able to back it up; she, of course, would be able to.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.